ARTICLE 2871 (More Doctors) Doctors) (More DOI: 10.1590/1413-81232015219.16622016 ------Mais Médicos Mais Médi Mais Program. In 2014, 2014, In Program. . imary Health Care, Income In Income imary Care, Health Mais Médicos Mais Pr This article an adap propose aims to (More Doctors) Program. The indicators were The indicatorswere Doctors) Program. (More 1 significant. However the averages of the Howeveraverages significant. evaluated the indicators better were in the group of municipa lities characterized throughAPS having the best the hi relation the income to indicator, In scores. ghest of amount money spent per capita in health APS performance the best to is related in the mu adaption ofthis methodology An nicipalities. may providebe to able a better understanding of the policies health care to related words Key Evalua Provision, Medical equality and Health, Systems tion of Health Shy (2002) to assess (2002) to the quality in of health care Shy the municipalities which joined the cos eachadapted one of for the nine criteria proposed in the methodology medium applied and were to and large municipalities Re in the Metropolitan and after they before Alegre in were gion of Porto tegrated in the the municipalities groups grouped were three into analysis concerning An their to scores. according correlationsany betweenthe different sco group the municipalities for res and the health indica reveal evaluated,tors that were did not anything Abstract Abstract tation of the methodology used and Starfield by Strengthening Primary municipalities in the Strengthening Care Health , Alegre, of Region in the Metropolitan Porto introduction the ofafter the comparison an inter-municipal Program: 1 1 Av. Francisco Francisco Av. Serviço Saúde de 91350-200 RS Alegre RS 91350-200 Porto [email protected] Brasil. Comunitária, Hospital Hospital Comunitária, Senhora da Nossa Conceição. Redentor. Cristo 596, Trein 1 Margarita Silva Diercks Silva Margarita Kopittke Luciane Claunara Schilling Mendonça Claunara Mendonça Schilling 2872 et al.

Introduction was to use the aforementioned methodology as Mendonça CS Mendonça a starting point to make subsequent comparisons Based on the objective of dealing with the lack between the municipalities and regions and to of doctors in priority regions that work in SUS search for the cutting off point for the score re- and to reduce regional inequalities in health care, sults that would mean an alteration in the quality the Federal Government set up the Mais Médi- of primary health care using the classic indicators cos (More Doctors) Program (hereafter PMM) such as: post-neonatal mortality rates, reductions in Brazil. This Program is part of a wider project in admissions and avoidable deaths. aimed at improving the service given to users of the Brazilian National Health Service (hereafter SUS) that also involves: more investment in basic Methods health centers, increasing the numbers of med- ical school vacancies to train more doctors and The data used to assess the operational effective- having more medical residencies to improve the ness of the APS in the municipalities took into quality of teaching for doctors in Brazil, all of account macro characteristics which are indica- which are directed towards Primary Health Care tive of a Health System based on primary care. (hereafter APS)1. The characteristics of an APS service are not The metropolitan region of Porto Alegre is being considered in this paper as there are other made up of 12 medium and large municipalities studies that will provide this type of analysis us- in the region, and has 31.6% of the state’s popu- ing tools such as PCATool to do this. lation. Before the implementation of the PMM The original criteria used to classify Health for primary health care, this area had one of the Systems guided by APS that was used by Starfield lowest coverages for doctors in the country, be- included: extending regulations for the health ing 20.7% and two of the municipalities did not system from the perspective of APS, the type of even have one designated medical team until APS professional, the percentage of active doc- 2012. The objective of this analysis was to use the tors that are specialists, APS doctors earnings in criteria created by Starfield and Shy2, which was relation to specialists, a division of APS costs, used to compare the operational effectiveness of the registration and list of patients, the existence 13 primary health care centers in industrialized of 24 hour service coverage and the importance countries and then adapt and apply them to the of family health care being covered in medical 12 municipalities in the metropolitan region of school courses2. Porto Alegre/RS. The purpose was to measure the In order to classify the positions and states of characteristics of primary health care and their APS for each municipality, a points based criteria relations with the level of health care in the pop- was developed based on 8 characteristics of health ulation before and after the implementation of systems. They were adapted for use at local levels PMM. 244 doctors were working in these areas in Brazilian municipalities. Each characteristic in 2013. was given a numerical point from 0 (not having it According to Starfield, having a large num- or very poor performance in relation to this char- ber of doctors working in the primary health acteristic) to 2 (high level of development in this care network will: reduce the adverse effects of characteristic). A point of 1 was given where the social inequality, increase the amount of doctors development of the characteristic was considered available in the primary health care network and to be moderate. All the points were added up and increase the number of available specialists doc- the averages were taken, in order to obtain the tors. This will result in better health care indica- total scores for the primary health care networks tors for the population3. In studies carried out in in each municipality for 2012 covering the intro- developed countries, the findings showed that duction of the PMM and then in 2014 after the good APS meant lower costs in the general health program was implemented (Table 1). care system which in turn has a beneficial impact The indicators that were adapted for each of on the health of the population4. The adaption the 8 criterion are below: of this tool to evaluate the operational effective- Criterion 1: Coverage of Family Health Strat- ness of APS at a infra-national level, in this case, egy. In order to characterize the regulation of the was relative since from the outset there were low municipal health system through APS, the popu- coverage levels for APS and a high level of social lation ought to access the health system services vulnerability which was the reason why the PMM from the services or professionals in APS. This was implemented in the region. The proposal criterion takes into account the population cov- 2873 Ciência & Saúde Coletiva, 21(9):2871-2878, 2016

Table 1. Scores covering the characteristics of the Health System related to the operational effectiveness of the primary health care network and total scores, 2012 and 2014.

Criteria 1 2 3 4 5 Year 2012 2014 2012 2014 2012 2014 2012 2014 2012 2014 1 1 0 0 1 1 1 1 1 1 0 1 0 0 0 0 0 0 1 1 0 1 1 2 0 1 1 1 1 2 0 1 0 0 0 1 2 2 1 1 Gravataí 0 1 0 1 1 2 1 1 2 2 Guaiba 0 0 0 0 0 0 0 0 1 1 0 1 1 1 1 2 1 1 1 2 Porto Alegre 0 1 1 1 1 1 0 0 1 2 0 0 0 0 0 2 1 1 0 1 0 1 0 1 1 2 2 2 1 1 São Leopoldo 0 0 1 1 0 1 1 1 1 1 Viamao 0 0 0 0 0 2 1 2 0 1

Criteria 6 7 8 Score Score Year 2012 2014 2012 2014 2012 2014 2012 2014 Alvorada 0 1 0 0 0 0 2,5 3,1 Cachoeirinha 0 1 1 1 0 1 1,2 3,1 Canoas 2 2 2 2 1 2 5 8,1 Esteio 0 1 0 0 1 1 2,5 4,3 Gravataí 2 2 2 2 1 1 5,6 7,5 Guaiba 1 1 0 0 0 0 1,2 1,2 Novo Hamburgo 2 2 2 2 0 2 5 8,1 Porto Alegre 2 2 2 2 1 2 5 6,8 Sapiranga 0 1 0 0 0 1 0,6 3,7 Sapucaia do Sul 2 2 2 2 1 2 5,6 8,1 São Leopoldo 1 1 0 2 0 2 2,5 5,6 Viamao 0 1 1 2 0 1 1,2 5,6

1. Type of System; 2. Financing; 3.Type of APS Professional; 4. Percentage of specialists doctors; 5. co-payment by patients in the primary health care network; 6. Register and/or list of patients; 7. coverage 24 hours and 8, The strength of the academic departments in Family Medicine.

erage for Family Health (between 0 and 40% = tors from the health strategy for families and 0, between 41 and 60% = 1 and above 60% = 2). family/community health center doctors. The Criterion 2: Public resources per capita. This scores were given based on the ratio of doctors criterion deals with financing in the health sys- per inhabitant compared with the total amount tem in Brazil which is a mix system with part of of doctors in outpatient units in the municipal- the population being catered for by the private ity. Ratios considered less than 0.1 = 0, between sector whilst the majority rely on financing from 0.15 to 0.29 = 1 and > 0.3 = 2. tax revenues, which leaves the system underfund- Criterion 4: Percentage of active doctors that ed. For the financing score we used the figure of were specialists. The number of specialists were expenditure per capita in health. We then dis- considered to be all those other doctors in the counted all those from the population who were APS, outpatients units and policlinics taken from served through supplementary health. If it was the total amount of general practitioners in the less than R$ 500.00 = 0, between R$ 501.00 and aforementioned medical areas. Above 70% = 0, 1,000.00 = 1, and more than R$ 1,000.00 = 2. from 51 to 69% = 1 and < 50% = 2. Criterion 3: Type of APS Professional. Ac- Criterion 5: Co-payment in APS. This crite- cording to the data from the National Register rion entered into the scores as it substituted the of Health Establishments, the following doctors salary relations of APS doctors and specialists needed to be considered: doctors from APS, doc- doctors because we did not have any sources of 2874 et al.

information to undertake these calculations. In dicators, the low weight at birth was used as well Mendonça CS Mendonça the Brazilian model of primary health care there as: the infant mortality rates, the rate of avoid- is, in theory, no co-payment for services. Howev- able mortalities in those less than 5 years old, er, due to low investment in health care in some the rate of premature mortality for those with municipalities, patients often have to pay from chronic, transmissible diseases for those who their own pockets for essential medication. In were between 30 and 69 years old and the rate order to evaluate co-payment, we used the per of mortality due to External Causes. Information capita value of pharmaceutical assistance to get on the rate of admissions for sensitive conditions an idea of the expenditure from peoples own for APS was considered. All of the above was ob- pockets. If the expenditure per capita, taking into tained from the Health Ministry’s Information account Pharmaceutical Assistance was less than System9. R$ 5.00 = 0, from R$ 5.1 to 10.00 = 1 and greater Pearson’s correlation coefficient was used than 10 = 25. to measure the inter-correlations between the Criterion 6: List and register of the patients. effectiveness scores for the primary health care We considered having a defined area according networks and the distribution of expenditure in to the information from the Supervision of the health. Also Spearman’s correlation was used to PMM Report whose sources were tutors from the measure the relation between APS rankings in municipalities. No area defined = 0, having an the municipalities and the health indicators. The area that was defined and registered with a num- level of significance was determined as p < 0.05. ber of inhabitants greater than desired for doc- tors/team = 1 and having a defined and registered population and an information system = 2. Results Criterion 7: Coverage 24 hours. We consid- ered whether there were excellent Accident and In Table 1 are the score figures for each criterion Emergency (A & E) and Primary Health Care and the total scores for 2012 covering before the Units in the municipalities. If there were no 24 PMM and 2014 after the PMM was implement- hour services in the municipality = 0. If the gen- ed. We noted that before the PMM the scores eral hospitals covered A & E and primary health were very low in these municipalities, where only care services but this coverage was insufficient for two groups had been formed. The average total the entire population in the municipality = 1. If scores were 1.8 amongst 7 municipalities and 5.2 there were decentralized emergency services and in the other 5 municipalities giving a total of 16 other important services were decentralized with points. Later in 2014 the municipalities could be reference to the use of APS teams = 2. grouped into three groups with average scores Criterion 8: Medical Schools with Depart- being 3, 5.6 and 7.7. The improvements in the ments/Professors from the area of Family and scores were principally due to: the increase in Community Health Care (hereafter MFC). This coverage of Family Health (criterion 1), a greater criterion was included due to its importance and amount of doctors in APS (criterion 2), an in- evidence obtained in other contexts6. If there crease in spending in pharmaceutical assistance were no medical school students in the APS = (criterion 5), the demarcation of the population 0. If there were experiences of students having to receive APS services (criterion 6) and widen- worked in APS units in the municipalities be- ing the services network through Accident and coming professionals = 1. If there were obvious Emergency units (criterion 7). This was similar examples of those becoming qualified in MFC at to what occurred in the medical schools with Universities using the municipal services as an increases in residencies in MFC and using APS area for becoming medically qualified = 2. services in the municipalities for training those The purpose was to evaluate the relation in the medical profession (criterion 8). between income inequalities and the levels of Table 2 shows the final scores based on the health amongst the different municipalities. The criteria used to evaluate the operational effec- idea was that reductions in inequalities would be tiveness of the primary health care network in the prerequisite for the allocation of doctors in the municipalities that were analyzed one year the poorest municipalities as the distribution of after the arrival of the first and second phases of income was based on the Gini Coefficiency for the Mais Médicos Program in 2014. The munic- 20107. All of the expenditure was used on health ipalities were grouped into three similar groups and the spending per capita was on health care in in relation to the results of the final scores and the municipalities8. In relation to the health in- covering expenditure per capita on health in 2875 Ciência & Saúde Coletiva, 21(9):2871-2878, 2016

these municipalities in 2014. Gini’s coefficient the municipality groups characterized by having was used corresponding to the year 2010 and the better APS scores compared with the groups hav- Gross Domestic Product per capita for the year ing the worse scores for avoidable mortalities for 2012. the under 5s. In Figure 1 we noted that spending values per Our analysis did not show anything signifi- capita that are very low, were associated with the cant when we analyzed the differences between municipalities with the worse APS scores. The the municipality groups and the results of the highest amount of spending per capita in health is related to the best APS performances (r = 0.74, p < 0.001) in these municipalities. This is in ac- cordance with the last studies in relation to Eu- ropean countries which showed a direct correla- tion between good primary health care and the 8,00 Sapucaia do Sul Novo Hamburgo Canoas health of the population. This resulted in fewer Gravataí unnecessary hospitalizations and relatively low Porto Alegre socioeconomic inequalities. Overall spending on 6,00 health care was higher in the countries with the Viamão São Leopoldo best APS structures10. APS Scores 4,00 Esteio An analysis of the correlations between the Alvorada Cachoeirinha three different groups of scores from the mu- Sapiranga nicipalities and the health indicators that were 2,00 evaluated did not show anything significant. Guaíba However, as can be seen in the Table 3, the per- 0,00 500,00 1000,00 1500,00 2000,00 centage indicators for: low weight at birth, in- fant mortality rates, rates of premature deaths Expenditure per capita in health care in 2014 by DCNT and the rates of ICSAP are better in municipalities with good scores. The averages in Figure 1. Scores for the Primary Health Care Network à the indicators that were evaluated were better in Health and Expenditure per capita in Health, 2014.

Table 2. Municipalities grouped by APS scores and economic indicators: Expenditure per capita in health, GDP per capita and Gini Coefficient*.

The Primary Health Care Total Scores Expenditure pc Gini GDP per capita Scores 2014 in Health (R$) coefficient R$ Low APS Scores Alvorada 3.1 261.70 0.44 8.599 Cachoeirinha 3.1 436.73 0.44 37.455 Esteio 4.3 563.74 0.48 33.491 Guaíba 1.2 526.02 0.47 27.709 Sapiranga 3.1 412.89 0.41 19.535 Intermediate APS Scores Viamão 5.6 237.30 0.48 10.410 São Leopoldo 5.6 1,217.39 0.53 21.049 The Highest APS Scores Canoas 8.1 1,689.33 0.51 45.501 Gravataí 7.5 787.87 0.45 26.767 Novo Hamburgo 8.1 1,046.82 0.53 24.385 Porto Alegre 6.8 1,854.30 0.61 33.883 Sapucaia do Sul 8.1 405.57 0.45 18.000

Source: The Reality Observatory and the Public Policies of the Vale do Rio in Sinos-Sinos observed; http://www.fns.saude.gov.br/; http://siops-asp.datasus.gov.br/cgi/siops/serhist/MUNICIPIO/indicadores.HTM * Gini Coefficient the level of concentration of income in specific population groups varies 0 to 1 where 0 is the total distribution of income (everyone in the population that have the same income) and 1 means the total concentration of income (only one person has an income). 2876 et al.

Table 3. Averages for the Indicators of children’s health, mortality for Chronic non-transmissible Diseases Mendonça CS Mendonça in adults between the ages of 30 and 69 years old, mortality rate due to external causes and rates of Hospital Admissions for ambulatory care sensitive conditions in the Primary Health Care Network, for municipalities grouped by their APS scores.

The Primary Health % BPN MI Rate Rate of Avoidable Rate of Premature Death Rate Rate of Care Scores Deaths <5 years Deaths DCNT External Causes ICSAP Baixo Escore 9.02 10.54 10.69 410.03 429.63 28.02 Alvorada 8.98 9.80 11.3 465.31 468.14 30.85 Cachoeirinha 8.74 10.45 6.72 370.19 406.42 22.96 Esteio 8.22 13.40 9.83 389.62 445.32 34.91 Guaíba 9.49 9.60 10.38 465.60 477.63 27.69 Sapiranga 9.70 9.47 15.26 359.87 350.67 23.7 Médio Escore 9.82 10.45 9.43 443.22 373.19 26.26 São Leopoldo 9.77 12.29 8.09 348.55 422.12 23.48 Viamão 9.87 8.79 10.78 397.84 464.32 29.04 Maior Escore 8.56 9.12 12.23 436.64 393.63 24.44 Canoas 8.22 10.30 10.87 431.22 469.61 26.03 Gravataí 8.52 7.52 9.47 412.27 428.63 18.33 Novo Hamburgo 8.66 9.45 11.03 353.06 418.98 21.91 Porto Alegre 9.26 9.47 11.50 371.23 415.00 31.94 Sapucaia do Sul 8.15 8.89 18.30 400.38 451.00 24.02

Source: MS/SVS/DASIS. Information System on those born alive.Information System on mortality. Datasus.gov.br/Indicadores de Saúde.

health indicators. Nevertheless one can see a ten- ipalities with the lowest rates of doctors per in- dency where the best APS scores are associated habitant where there is extreme poverty and high with the best health indicators. necessity rates for health care. There was a reduc- tion by 53.5% in the number of municipalities with a scarcity of health care13-15 professionals in Discussion the year after the program was launched15 . It also brought to light the deficiencies in the infrastruc- 20 years ago there was the creation of the Brazil- ture and the organization of the services geared ian APS model that we know as Family Health towards meeting the needs of the population. in the metropolitan region of Porto Alegre which When we analyzed the municipalities in RM covers 40% of the population in the state of Rio in Porto Alegre, the results from the indicators Grande do Sul. At this time it started with the suggested different stages in the organization of development of an organized health care system primary health care even though they were in no based around primary health care for the pop- way significant when we also analyzed the differ- ulation. Innumerable international studies have ences between the municipality groups. This was shown that better health indicators in health care because the intermediate stages values were very systems occur where they are based on the guid- close to those that had low scores. In the munici- ing force of APS. The studies have shown better palities with lower populations linked to APS, the cost benefits even in regions with high levels of indicators were worse, whereas the municipali- inequalities11,12. This was not the case in the re- ties with relatively stable access to APS showed gion that we studied. The program for bringing better scores and better health indicator results. in more doctors to work in APS services in the There were some potential limitations with state of RS which started in 2012 with PROV- this type of analysis. These include: using indica- AB and then in 2013 with the PMM, brought to tors that only cover one year after the implemen- light the needs of the municipalities that were ei- tation of the PMM, subjectively analyzing the ther starting to or continuing in reforming their score criteria, the use of secondary data that may health systems through bolstering their APS. be limited with reference to what they register (or The PMM has clearly shown the need for more not showing results that we had not found) and doctors in the country particularly in the munic- a lack of specific indicators that measures the ef- 2877 Ciência & Saúde Coletiva, 21(9):2871-2878, 2016

fectiveness of the APS services where the doctors In the context of the PMM which Brazil were placed. Out of indicators that are normally intends to fully introduce through APS in the used to evaluate health care systems only some of municipalities, it is important to focus on the them reflected good primary health care, which aspects in the health care systems that are more was the case for post-neonatal mortality rates. related to the best health care results for the pop- These rates could not be analyzed in detail this ulation. If, in other studies in different countries, study as the data from the Mortality System was the operational effectives of the APS is related to not available for the year 2014. how much is spent in health care, this study has In relation to the financing criterion, areas demonstrated that a minimum spending thresh- designated for cuts were underestimated if we old exists that needs to be surpassed in order to consider what is adequate financing for APS obtain better indicators. Out of the 8 character- based on international examples or using good istics that we evaluated, upon using the origi- examples from Brazil. In large municipalities that nal study from the OECD, three of them made had the groups with the best scores, financing of a difference in countries showing low scores for APS competed with financing of other speciali- APS in two areas: universal financing/the equi- ties in the large third party hospitals and A & E table distribution of resources and the absence units. We suggest that expenditure per capita in of payment by users of the APS services. We can primary health care in relation to any remaining therefore use this data that is available, which is expenditure should be done in subsequent evalu- important, in order to establish priorities and ations. Bearing in mind the criteria in the current implement necessary changes. scenario for the evaluated municipalities, they The health services can contribute to reduc- would have scored better for diagnosis, admis- ing inequalities in health care10,11 particularly sions and emergency services as nearly all of the when these services are needed in shaping the spending in these services in these areas would be primary health care networks in the municipal- covered in primary health care (as is the case for ities. It is therefore necessary to use a methodol- Cachoeirinha and Esteio). ogy that monitors these scores with the inclusion Including analysis of the health care sys- of other indicators that are more specific for APS tem indicators for income (GDP per capita and services in order to evaluate to what extent the spending per capita in health) and inequalities municipalities are following the APS. (Gini rates), they reinforce the importance in us- The creation of a methodology that permits ing these indicators in any analysis on the evalu- comparisons between and inside of municipali- ation of health systems. The majority of munici- ties over a period of time, in spite of their limita- palities with a Gini rate of 0.4 that were not a part tions, have become fundamental in demonstrat- of the group with the lowest APS scores, showed ing the importance of continuity in investments low income inequalities, however, due to the for the achieving of better organized APS services GDP per capita the other rates were also very low. and the coordination of primary health care net- In other words, low rates do not necessarily mean works for the population. This is principally in that these municipalities showed good economic the context of major social inequalities as is the development. There was a lack of any correla- case in the metropolitan region of Porto Alegre. tions when the municipalities were considered to be very poor performers. The importance of this aspect in our analysis is that the PMM with Federal funding for these municipalities took the onus from the municipalities to spend in health care which is important. Where there are eco- nomic crises, one can infer that the absence of Collaborations this program would leave the municipalities even more vulnerable in relation to access to medical CS Mendonça participated in coming up with care and for having good health care outcomes the idea for this paper as well as the methodolo- for the wider population. Subsequent studies in gy that was used and drafting the final version. L this area should look at the impacts of the above, Kopitke participated in designing and complet- comparing these municipalities with others that ing the database and obtaining information for do not have the program and which have similar this study. MS Diercks reviewed the final version economic vulnerabilities. of this paper. 2878 et al.

References Mendonça CS Mendonça

1. Brasil. Presidência da República. Casa Civil. Lei nº 11. Mackenbach JP. An Analysis of the Role of Health Care 12.871, de 22 de outubro de 2013. Institui o Programa in Reducing Socioeconomic Inequalities in Health: Mais Médicos, altera as Leis no 8.745, de 9 de dezembro The Case of the Netherlands. Int J Health Serv 2003; de 1993, e no 6.932, de 7 de julho de 1981, e dá outras 33(3):523-541. providências. Diário Oficial da União 2013; 23 out. 12. Starfield B, Shi L, Macinko J. Contribution of Primary 2. Starfield B, Shi L. Policy relevant determinants of Care to Health Systems and Health. Milbank Q 2005; health: an international perspective. Health Policy 2002; 83(3):457-502. 60(3):201-218. 13. Oliveira FP, Vanni T, Pinto HA, Santos JTR, Figueiredo 3. Starfield B. New paradigms for quality in primary care. AM, Araújo SQ, Matos MFM, Cyrino EG. Mais Médi- Br J Gen Pract 2001; 51(465):303-309. cos: um programa brasileiro em uma perspectiva in- 4. Saltman RB, Rico A, Boerma W, editors. Primary care ternacional. Interface (Botucatu) 2015; 19(54):623-634. in the driver’s seat? Organizational reform in European 14. Santos LMP, Costa AM, Girardi SN. Programa mais Primary Care. Berkshire: McGraw-Hill; 2011. Médicos: uma ação efetiva para reduzir iniquidades em 5. Barros AJD, Bertoldi AD. Out-of-pocket health expen- saúde. Cien Saude Colet 2015; 20(11):3547-3552. diture in a population covered by the Family Health 15. Estação de Pesquisa de Sinais de Mercado (EPSM). Program in Brazil. Int J Epidemiol 2008; 37(4):758-765. Dados Estatísticos sobre o impacto do Programa Mais 6. Giovanella L. A Atenção primária à saúde nos países Médicos no cenário de escassez de médicos em atenção da União Europeia: configurações e reformas organi- primária no Brasil. Belo Horizonte: EPSM/NESCON/ zacionais da década de 90. Cad Saude Publica 2006; FM/UFMG; 2015 [acessado 2015 set 29]. Disponível 22(5):951-964. em: http://epsm.nescon.medicina.ufmg.br/epsm/Pes- 7. Observatório da Realidade e das Políticas Públicas do quisa_Andamento/Impacto_Programa_Mais_Medi- Vale do Rio dos Sinos - ObservaSinos. [acessado 2015 cos.pdf set 29]. Disponível em: http://www.ihu.unisinos.br/ 8. Brasil. Ministério da Saúde (MS). Secretaria Executiva. Departamento de Economia da Saúde, Investimentos e Desenvolvimento. Sistema de Informações sobre Orça- mentos Públicos em Saúde (SIOPS). [acessado 2015 set 29]. Disponível em: http://www.siops.datasus.gov.br/ consExecFinanceira.php?Ano 9. Brasil. Ministério da Saúde (MS). Secretaria Executiva. DATASUS. Informações de saúde. Informações Epide- miológicas e morbidade. [acessado 2015 set 29]. Dispo- nível em: http://www.datasus.gov.br 10. Kringos DS, Boerma W, van der Zee J, Groenewegen P. Europe’s Strong Primary Care Systems Are Linked To Article submitted 19/03/16 Better Population Health But Also To Higher Health Approved 20/06/2016 Spending. Health Aff 2013; 32(4):686-694. Final version submitted 22/06/2016